Illustration created by Midjourney with the prompt “graffiti, spraypaint, colorful, vibrant, showing ai, tech, digital transformation, gadgets, ordinary things, street art –ar 3:2 –v 4”
Yes. This is another article about ChatGPT. But it is also career advice for those of you who love ChatGPT, hate it, fear it, think it’s a fad, and wish everyone would stop writing about it and those who are simply not paying attention. No matter where you fall on this spectrum, I want you to think about the following.
The commonly expressed thesis of Charles Darwin’s On the Origin of Species is “survival of the fittest.” But that’s not what Darwin meant, and it is clearly not the hypothesis behind the Theory of Evolution. A more accurate summary of both the book and the theory would be “survival of the most adaptable.” The evidence (across evolutionary time scales) suggests that species that are most able to adapt to ever-changing external forces survive to pass on their genes. Those that cannot adapt (no matter how “fit” they are) do not survive to pass on their genes.
The pace of technological evolution is orders of magnitude greater than the pace of biological evolution, but “survival of the most adaptable” presents an excellent framework for the allocation of your resources – which brings us back to ChatGPT.
Regardless of your opinion of ChatGPT or other recent generative AI applications, it’s time to get involved and learn more about them. Why?
Remember Bob, from back in the day? He wasn’t tech support. He was just a coworker a few cubicles down the hall who could fix your “blue screen of death” or get the printer to print. Even though it wasn’t his day job, Bob was passionate about computers and tech and everyone on the floor knew it – including the boss. Bob’s little tech side hustle made him an invaluable team member, and his expertise got him into conversations that led his career to places it otherwise could not have ventured.
It’s Early Days
The hype-cycle around generative AI is in high gear. I anticipate the creation of hundreds, if not thousands, of AI models and applications in the coming months, all competing for attention. I also expect most of these companies to disappear almost as quickly as they form. It will be hard to differentiate 10 different companies that are all offering meeting summary applications built on the same AI models.
There is also a distinct possibility that current approach to generative AI (specifically the approach being used by GPT-3, LaMDA, and the other large language models) will prove to be a “tourist attraction” on the way to a completely different approach to more capable artificial intelligence models.
Whether or not the hype is warranted and regardless of the underlying technology’s ultimate fate, you can create real value by learning some new skills.
Your Next Steps
First, get a ChatGPT account. Get a paid one if you can. It’s just $20/month. You pay monthly and you can cancel it at any time. Next, read some blog posts about getting the most out of the application. You’ll want to get serious about prompt crafting (aka prompt engineering or prompt tuning). Spend no less than 15 minutes using this software every day for the next two weeks.
After that, go get access to Midjourney on Discord. Pay for this as well (it’s a better experience). Again, you can cancel at any time. Explore the Midjourney Discord server to learn how to create artwork from text prompts. Spend no less than 15 minutes using this software every day for the next two weeks.
After spending seven-eight hours (2 weeks, ~15 minutes per day with each AI) working with ChatGPT and Midjourney, you are going to have a very strong personal opinion about these two approaches to generative AI. You may consider these tools invaluable AI-coworkers. You may consider them nonsense. What will be true is that the opinion you have will be based on your lived experience, not on someone’s blog post (like this) or someone’s YouTube video.
I Don’t Need It!
As you can imagine, many of the people who ask me about generative AI are only asking so they can tell me that they’ve given it a try and (1) it’s not accurate, (2) the writing wasn’t useful, or (3) they can’t really see a need for it. When I ask how much time they spent learning how to use the applications (specifically learning and practicing prompt crafting), the answer is either “none” or “a little bit.”
Forgive me for stating the obvious, but if you use the wrong end of the hammer, you’re not going to find it as useful as it could be.
What’s Next?
This will not be my last essay about generative AI applications, models, or infrastructure. These tools are not new, but they are news. We are at an inflection point. The extraordinary publicity around ChatGPT has the world focused on AI. Historically, public awareness at this level has been followed by significant transformation. So, the “smart money” says, it’s time to transform.
Author’s note: This is not a sponsored post. I am the author of this article and it expresses my own opinions. I am not, nor is my company, receiving compensation for it.